The present inventions relate generally to methods for processing signals obtained by transducers in subterranean wells. More specifically, the inventions relate to the use of wavelet analysis in improved methods of obtaining and processing signals relating to physical parameters downhole derived from downhole transducers.
It is known to measure various physical parameters of the subterranean environment of an oil or gas well. Physical parameters such as pressure and temperature are commonly converted into electronic signals with downhole transducers. Such signals can be analyzed for making decisions regarding the operation of the well.
One example of a common type of well test is known as a build-up test. In a build-up test a pressure transducer is introduced into the well and the well is temporarily sealed. The build-up of pressure is monitored and the information obtained is often used to make operational decisions relating to production and further development of the well.
Conventional systems, using for example quartz piezoelectric transducers, are subject to errors caused by static and dynamic pressure and temperature induced errors. No matter how well designed, the output signal or data from a quartz pressure transducer is affected by the temperature of the pressure sensing piezoelectric element, and the data from a temperature sensing quartz element is effected by pressure. It is common to calibrate (correct) pressure data as a function of temperature data collected at or near the pressure-measuring site. Substantial computation is required in order to convert and correct the output into an intelligible form.
Complicating matters, oftentimes transducer signals can also be affected by interference caused by the operation of well tools, changes in the downhole environment by the opening or closing of valves, or a number of other causes. Using conventional analysis techniques it is often unknown whether a particular data signal is valid until a significant time interval has elapsed. Additionally, using conventional techniques, when data is determined to be flawed, it is often impossible to determine whether an entire data signal is flawed or only isolated portions of the data. Generally, it is impossible to extract useful information from data known to be flawed.
Despite recent advances in the making and use of downhole transducers to gather data relating to physical parameters relating to the downhole environment, due to the above-mentioned problems with current data collection and analysis techniques, attempts to measure downhole conditions sometimes fail. For example, with state-of-the-art equipment and methods, it is possible to conduct a pressure build-up test in its entirety before discovering upon further analysis that the test was flawed. Much time and expense can be wasted in analyzing and reproducing tests before errors are discovered and in retesting after the discovery of errors.
Generally, the present invention is directed to providing improved methods of analysis of downhole test data capable of providing almost instantaneous indications of not only the test parameters, but also the presence or absence of error conditions. The new analysis methods provide a useful verification of well data collection efforts. The new analysis methods also provide new alternative methods of recording or reconstructing well data. The real-time aspect of the invention is extremely advantageous in that it provides feedback useful for real-time correction of error conditions.
The invention provides methods for combining well test analysis with wavelet analysis. The use of wavelet analysis with a conventionally acquired well data signal provides verification of the well data signal. The methods of the invention include the steps of acquiring downhole data, converting the data to a first electronic signal, performing wavelet analysis of the first electronic signal to produce a second electronic signal, and using the second electronic signal to aid in the interpretation of the first electronic signal and to instigate physical feedback into the well environment.
According to one aspect of the invention, the wavelet analysis is performed using a Daubechies wavelet.
According to another aspect of the invention, the wavelet analysis is performed using a Daubechies 10 wavelet.
According to still another aspect of the invention, the wavelet analysis is performed using a Daubechies 1 wavelet to produce the first derivative of the first electronic signal with respect to time.
According to another aspect of the invention, the wavelet analysis is performed in real-time.
According to yet another aspect of the invention, all of the steps are performed in real-time.
According to other aspects of the invention, based upon the information contained in the second electronic signal, steps are taken to influence physical conditions in the well environment, subsequently affecting the first electronic signal.
According to an additional aspect of the invention, the added step of reconstructing the first electronic signal from the second electronic signal is performed.
According to other aspects of the invention, the methods are employed to measure downhole well pressure.
According to other aspects of the invention, the methods are employed to perform well tests such as build-up, interference, drawdown or pulse tests.